Recent studies have shown that TGF-B inhibits cell proliferation by blocking cell cycle progression at the G1 phase of the cell cycle and hence, is thought to function as a tumor suppressor protein. Most human cancers appear to have lost their growth-inhibitory response to TGF-B. Our preliminary studies show that foregut cancers with inactivation of TGF-B signaling express high levels of cyclin D1 and CDK4 levels, suggesting that the deregulated expression of these proteins may contribute to the development of these tumors. Flavopiridol, an established Cdk inhibitor used in treatment protocols for certain cancers, is known to inhibit most known CDKs especially against CDKs 7, 8 and 9, is thought to be largely responsible for the toxic side effects caused by this drug in clinical trials. In light of these observations, we have screened a compound library of kinase inhibitors for CDK-4 specific inhibition and isolated two novel molecules (ON55290 and ON27900) which exhibit CDK4 inhibitory activity. In this application, we propose to carry out pre-clinical studies to test the usefulness of these molecules in gastrointestinal cancer therapy. The aims of the proposal are: 1. To further validate the role of CDK4 in Elf+/-, Elf+/-:Smad3+/- tumor model system through evaluation of tumor incidence in CDK4+/V El+/- : CDK4+/-: Elf/Elf[+]-Smad4[+/-] and CDK4[+/-]; Elf/Elf+:Smad4+/- mice. 2. (a) To expand the chemical library in an effort to understand the structure-activity relationship (SAR) of CDK4 inhibitors and (b) to conduct a detailed kinetic analysis of CDK4 inhibition by ON55290 and ON27900 to gain critical information on the mechanisms by which these compounds elicit their inhibitory effects on CDK4. 3. To determine the molecular mechanisms by which ON55290 and ON27900 bring about growth arrest and death of human gastrointestinal cancer cells. 4. To assess the safety and pharmacokinetics of the candidate drug in animal models of gastrointestinal cancers that develop in the TGF-f3 inactivated state.